User’s Guide to CIRFLE vs 2.0
CONCEPTUAL IRRIGATION RETURN
FLOW HYDROSALINITY MODEL
Prepared by:
D. Quílez, D. Isidoro, R. Aragüés
(CITA – Government of Aragón, Spain)
March 2009
INDEX
1. Introduction
2. Mathematical description
2.1. Hydrological submodel
2.2. Salinity submodel
3. Installing CIRFLE
4. Running CIRFLE
4.1. CIRFLE application
4.2. “Project” menu
4.3. “Area” menu
4.4. “Input Data for the Area” menu
4.5. “Output Data for the Area” menu
4.6. “Sensitivity Analysis” menu
4.7. “Irrigation Project Input” and “Irrigation Project Ouput” menus
5. Description of Inputs
5.1. Inputs defining the water balance
5.2. Inputs defining salt loading and salt leaching
5.3. Inputs defining salinity in the input water fluxes
5.4. Conversion factors for estimating TDS from EC
6. Description of Outputs
6.1. Water and salt fluxes
6.2. Irrigation performance and salt loading parameters
7. References
1. INTRODUCTION
CIRFLE is a computer model that estimates the volume of water and the salt
concentrations and loads in irrigation return flows. The model CIRFLE focuses on the crop's
root zone and considers only the main flow-paths of water and salts in the system. The model
assumes that the masses of water and salt are conservative and that steady state conditions
can model long-term transient conditions approximately.
CIRFLE is based in a mass balance approach for water and salts, and considers only the
most important inputs and outputs from the system. Inputs to the system are irrigation,
precipitation, and inflows from rim areas; outputs from the system are evapotranspiration,
irrigation and precipitation runoff, subsurface drainage and deep percolation. The changes in
soil storage are considered as the difference between the initial and final states, with special
attention to the leaching efficiency of salts and the calcite and gypsum dissolution-precipitation
processes. CIRFLE does not model individual ions, only salts as a lumped parameter, and thus
cation exchange reactions or adsorption are not considered.
The model has been designed to be applied to large systems and for long periods, such as
an irrigation season, a hydrologic year, or a series of consecutive years. The model should not
be applied for short periods as, in general, steady state conditions do not hold. The model's
variables and parameters are spatial and temporal averages. Thus, CIRFLE must be applied to
spatially homogeneous areas or the results can be misleading.
The conceptual Irrigation Return Flow hydrosalinity model was developed by Tanji (1977),
revised by Aragüés et al. (1985, 1990) with the name CIRF, and updated by Quílez (1998) with
the name CIRFLE (Conceptual Irrigation Return Flow hydrosalinity model with consideration for
the Leaching Efficiency of salts). The model has been revised, updated and rewritten in the
object-oriented programming language C-sharp (C#).
2. MATHEMATICAL DESCRIPTION of CIRFLE
CIRFLE consists of a hydrologic submodel coupled to a salinity submodel. In the
hydrologic submodel the volume of water Q is considered. In the salinity submodel, the salt
concentration (C) expressed as total dissolved solids (TDS), and the load or mass of salts (M)
are considered. Salt load is obtained as the product of water volume, salt concentration and an
adequate unit conversion factor (SMCF) that depends on the units of input data.
2.1. Hydrologic submodel
Figure 1 shows the inputs, outputs and flow pathways considered in the hydrologic
submodel. The mathematical equations are presented in Table 1 and the definition of symbols
in Table 2.
The inputs to the system are diverted irrigation water (Qdiw), precipitation (Qp) and rim
inflows from lateral systems (Qrim) (Eq. 1). The volume of irrigation water that evaporates directly
from the soil surface or the plant canopy (Qevdiw) is defined through the irrigation water
evaporation coefficient (iwec) (Eq. 4), given as an input to the model. The volume of water that
Figure 1. Diagram of the hydrologic submodel, focusing on crop root zone. The symbol
Q denotes quantity of water.
effectively infiltrates the soil or “effective applied irrigation water (Qeaiw)” is defined through the
coefficient Eiae, the irrigation application efficiency (Eq. 5). The rest of Qdiw that does not
infiltrates the soil or evaporates is regarded as the volume lost as surface runoff (Qiwro) (Eq. 6).
Precipitation (Qp) follows the same pathways than irrigation: precipitation evaporation
(Qevp), precipitation runoff (Qpro) and effective precipitation (Qep) (Eq. 8, 9 and 10, where pec =
precipitation evaporation coefficient and prc = precipitation runoff coefficient).
The soil water available for evapotranspiration (Qsw) is obtained by adding the effective
irrigation water (Qeaiw), the effective precipitation (Qep) and the initial stored soil water (Qisw) (Eq.
14).
The evapotranspiration (Qet) is considered at this point, being Qpsw the volume of water
remaining in the soil after evapotranspiration. The evapotranspiration concentration factor
(ETCF), defined as the ratio Qsw / Qpsw (Eq. 16), indicates the concentration factor of the soil
water due to ET. The volume of water after ET (Qpsw) is decomposed in its three components:
the effective applied irrigation water after ET (Qsweaiw), the effective precipitation after ET (Qswep)
and the initial soil water after ET (Qswisw) (Eq. 15). The final stored soil water (Qfsw), an input to
the model, is subtracted from Qpsw to obtain the water available for subsurface drainage and
deep percolation (Qppsw) (Eq. 17).
Table 1. Mathematical equations describing the hydrologic submodel. Q denotes volume
of water.
Hydrological inputs and outputs
(1) Qi = Qdiw + Qp + Qrim
(2) Qo = Qiwro + Qpro + Qet + Qevdiw + Qevp + Qdp + Qsdw + Qrim
Diverted irrigation water
(3) Qdiw = Qeaiw + Qevdiw + Qiwro
(4) Qevdiw = evdiw · Qdiw
(5) Qeaiw = Qdiw · Eiae
(6) Qiwro = Qdiw · (1 – evdiw - Eiae)
(7) Qsweaiw = Qeaiw / ETCF
Precipitation
(8) Qp = Qpro + Qevp + Qep
(9) Qevp = pec · Qp
(10) Qpro = prc · Qp
(11) Qep = (1-prc-pec) · Qp
(12) Qswep = Qep / ETCF
Initial stored soil water
(13) Qswisw = Qisw / ETCF
Water inputs to root zone
(14) Qsw = Qeaiw + Qep + Qisw
(15) Qpsw = Qsw – Qet = Qsweaiw + Qswep + Qswisw
(16) ETCF = Qsw / Qpsw
Final stored soil water
(17) Qppsw = Qpsw – Qfsw
Water outputs from root zone
(18) Qdp = dpc · Qppsw
(19) Qsdw = (1 – dpc) · Qppsw
(20) Qsw = Qet + Qfsw + Qdp + Qsdw
(21) Dl = Qppsw / A
Surface irrigation return flows
(22) Qsirf = Qiwro + Qpro + Qsdw + Qrim
Water use efficiency, leaching fraction
(23) WUE = Qet / (Qeaiw + Qep)
(24) LF = Qppsw / (Qeaiw + Qep)
Outputs below the crops root zone are the deep percolation (Qdp) and the subsurface
drainage (Qsdw). The deep percolation coefficient (dpc), an input to the model, gives the
percentage of the water available for subsurface drainage and deep percolation (Qppsw) that
percolates as Qdp (Eq. 18).
The surface Irrigation return flows (Qsirf) are the sum of the subsurface drainage (Qsdw),
irrigation runoff (Qiwro), precipitation runoff (Qpro) and rim inflows (Qrim) (Eq. 22).
Two additional equations calculate the water use efficiency (WUE) that gives the fraction
of the infiltrated irrigation and precipitation that undergoes evapotranspiration (Eq. 23), and the
leaching fraction (LF) that gives the portion of the infiltrated irrigation and precipitation that
percolates below the root zone (Eq. 24).
Table 2. Symbols used in the hydrologic submodel. Q denotes volume of water.
Symbol
Definition
A
Surface
Dl
Depth of water available for leaching
dpc
Deep percolation coefficient
Dr
Average rooting depth
Eiae
Irrigation application efficiency
ETCF
Evapotranspiration concentration factor
evdiw
Irrigation evaporation coefficient
LF
Leaching fraction
pec
Precipitation evaporation coefficient
prc
Precipitation runoff coefficient
Qdiw
Diverted irrigation water
Qdp
Deep percolation
Qeaiw
Effective applied irrigation water
Qep
Effective precipitation
Qet
Evapotranspiration
Qevdiw
Evaporation of irrigation water
Qevp
Evaporation of precipitation
Qfsw
Final stored soil water
Qi
Hydrologic inputs
Qisw
Initial stored soil water
Qiwro
Irrigation water runoff
Qo
Hydrologic outputs
Qp
Precipitation
Qppsw
Water available for subsurface drainage and deep percolation
Qpro
Precipitation runoff
Qpsw
Soil water after ET
Qrim
Rim inflow/outflow
Qsdw
Subsurface drainage water
Qsirf
Surface irrigation return flow
Qsw
Soil water before ET
Qsweaiw
Effective applied irrigation water after ET
Qswep
Effective precipitation after ET
Qswisw
Initial soil water after ET
SP
Saturation Percentage obtained in the soil saturation paste
WUE
Water use efficiency
2.2. Salinity submodel
Figure 2 shows the inputs, outputs and flow pathways considered in the salinity submodel.
The mathematical equations are presented in Table 3 and the definition of symbols in Table 4.
The inputs to the system are the salt concentration and salt mass in diverted irrigation water
(Cdiw, Mdiw), precipitation (Cp, Mp) and rim inflows from lateral systems (Crim, Mrim). In all cases,
the salt mass of each component is calculated from the product of the corresponding volume of
water and salt concentration and a unit conversion factor (SMCF) (Eqs. 29, 36 and 73).
To calculate the concentration of the effective applied irrigation water (Ceaiw) the increase
in the salt concentration of the irrigation water due to evaporation (1/(1-iwec)) is considered (Eq.
30 ). The salt mass in the effective applied irrigation water (Meaiw) is calculated as the product of
salt concentration (Ceaiw), water volume (Qeaiw) and the unit conversion factor (SMCF) (Eq. 31).
Irrigation can dissolve some of the salts present at the soil surface (Ciwrosp, an input to the
model) being Ciwro the final salt concentration of irrigation runoff (Eq. 32). Salt load in irrigation
runoff is obtained as the product of its concentration (Ciwro), the volume of irrigation runoff (Qiwro)
and the unit conversion factor (SMCF) (Eq. 33).
The TDS of precipitation (Cp) is concentrated based on the precipitation evaporation
coefficient (pec) to give the concentration of the effective precipitation (Cep) (Eq. 39). The salt
load in the effective precipitation (Mep) is calculated as the product of the volume of effective
precipitation (Qep), its concentration (Cep) and the unit conversion factor (SMCF) (Eq. 40).
CIFLE also considers that precipitation runoff (Qpro) can dissolved part of the salts present
at the soil surface (Cprosp), being Cpro the final concentration of precipitation runoff (Eq. 37). The
salt load in precipitation runoff (Mpro) is obtained from Eq. 38.
The electrical conductivity of the soil saturation extract (ECe), an input to the model, is
used to account for soil salinity. When there is “excess gypsum” in the soil, the ECe of the initial
soil water is corrected by gypsum solubility (ECgype) to give Cse by means of Eq. 46. “Excess
gypsum” in the soil is considered when the amount of gypsum in the soil (Gypsum, an input to
the model) is sufficient to saturate the soil water after ET (Qpsw). This correction is performed
because the TDS of a gypsum-saturated solution cannot be significantly concentrated during
evapotranspiration. To calculate the amount of gypsum potentially soluble in soil water the
3
model uses a gypsum solubility of 2.63 kg/m or 21 mM/L (Tanji, 1969) and a value of the EC at
saturation of 2.2 dS/m. Although these values are for deionized water and the approach has
limitations because gypsum solubility depends on other soil variables (Tanji, 1969), it is
preferred over that where it will be allowed to freely evapo-concentrate due to ET. If there is no
“excess gypsum” in the soil, then ECgype = 0 and TDSgyp = 0.
The corrected ECe for the “excess gypsum” condition is converted into TDS (Cse) using the
conversion factor of 1 dS/m = 720 mg/l of dissolved salts. This conversion factor was taken from
the approximate average for the surface waters of the Ebro river basin in Spain with similar
fractions of Cl, SO4 and HCO3 (meq/L) in the solution (CHE, 2006). The TDS of the initial soil
solution (Cisw) is obtained from Cse, and the ratio between water content at saturation (SP · Dr ·
A · Db) and the initial soil water content (Qisw) (Eq. 47). To avoid unreasonable low Cisw values in
cases where gypsum is the main source of salinity, the condition Cisw > Cmin was imposed,
where Cmin is the salt concentration of the soil solution in equilibrium with the salt concentration
of the irrigation water. The computation of Cmin is explained below.
The salt load of the initial soil water (Misw) is the product of the volume of the initial stored
soil water (Qisw), its salt concentration (Cisw) and the salt mass conversion factor (SMCF) (Eq.
49). The salt load in the soil water before ET (Msw) is calculated as the sum of the salt loads in
effective irrigation, effective precipitation and initial soil water (Eq. 55). The concentration of soil
water before ET (Csw) is obtained as the ratio between the salt mass (Msw) and the volume of
water (Qsw) and the salt mass conversion factor (SMCF) (Eq. 56).
Since crop’s evapotranspiration is free of salts, the salinity of the soil solution after ET
(Cpsw) is concentrated by the ET concentration factor (ETCF), but the salt load (Mpsw) remains
unchanged (Eq. 57). Cpsw is calculated as the ratio of the salt mass (Mpsw) to the volume of
water (Qpsw) and the salt mass conversion factor (Eq. 58).
As the soil solution is concentrated during the evapotranspiration process, some low
soluble minerals (as gypsum and calcium carbonate) may precipitate if their solubility products
are exceeded. Also, lime or gypsum can dissolve if they are readily available in the soil. CIRFLE
computes the term Msp - Msd (salt pickup – salt deposition, or mineral dissolution – mineral
precipitation) using the model Watsuit (Wu et al., 2009). Watsuit calculates the extent to which
applied irrigation water, as it becomes concentrated in the soil solution due to
evapotranspiration, effectively dissolves CaCO3 from the soil or precipitates out CaSO4·2H2O
and CaCO3. A linear regression equation that relates Csp – Csd with the leaching fraction (LF) is
given in CIRFLE to account for this effect (Eq. 52). Values of the intercept and slope of this
equation have been calculated for different irrigation water types with ionic compositions and EC
values considered representative of most irrigation districts. Nevertheless, the user may
introduce its own empirical values if desired. The salt mass due to salt pickup-salt deposition
(Msp-Msd) is obtained as the product of Qeaiw, (Csp - Csd) and the salt mass conversion factor (Eq.
53).
Salt pickup due to gypsum dissolution (Mgsp) in the “excess gypsum” scenario is now
considered. CIRFLE assumes a gypsum solubility of 2.63 kg/m3 and gypsum is dissolved until
the soil water after ET (Qpsw) is gypsum-saturated.
Adding the salt mass in the soil solution after ET (Mpsw), the salt pickup minus salt
deposition (Msp – Msd) and the dissolution of gypsum (Mgsp) gives the total mass of salts M’psw
(Eq. 59). The corresponding salt concentration (C’psw) is calculated as the ratio of M’psw to Qpsw
and the salt mass concentration factor (Eq. 60).
CIRFLE considers at this point the leaching efficiency of salts present in the soil using the
empirical approach developed by Hoffman (1986). The proportion of salts remaining in the soil
(Cf) with respect to the salts initially present (Ci) after a depth of water Dl has percolated through
a given depth of soil (Dr) is calculated as:
Cf - C min
k
=
C i C min D l + k
Dr
where Cmin is the salt concentration in the soil water after ET in equilibrium with the irrigation
and precipitation waters. Cmin considers the dissolution-precipitation of calcite and the
precipitation of gypsum in the soil (Csp-Csd) (Eq. 61). The empirical coefficient k takes into
account the inefficiency of salt leaching that, among others, depends on soil physical and
chemical characteristics, as pore size distribution, macropore bypass or soil water content.
Information on the parameter k is provided in section 3.1 (input variables).
Figure 2. Freebody diagram of salinity submodel, focusing on crop root zone. The
symbols C and M denote total dissolved solids and mass of salts, respectively.
The final concentration of soil water (C’fsw) is obtained from the initial concentration of soil
water using the leaching efficiency coefficient (k) and considering the amount of percolating
water (Qppsw/A) per unit depth of soil (Dr) (Eq. 62).
Table 3. Mathematical equations describing the salinity submodel
Salt inputs and outputs and change in storage
(25) Mi = Mdiw + Mp + Mrim
(26) Mo = Miwro + Mpro + Mdp + Msdw + Mrim
(27) Ms = Mfsw – Misw + Msd – Msp – Mgsp
Lateral contributions
(28) Mrim = Crim · Qrim · SMCF
Diverted irrigation water
(29) Mdiw = Cdiw · Qdiw · SMCF
(30) Ceaiw = Cdiw / (1 – iwec)
(31) Meaiw = Ceaiw · Qeaiw · SMCF
(32) Ciwro = Ceaiw + Ciwrosp
(33) Miwro = Ciwro · Qiwro · SMCF
(34) Csweaiw = (Msweaiw / Qsweaiw) /SMCF = Ceaiw · ETCF
(35) Msweaiw = Meaiw
Precipitation
(36) Mp = Cp · Qp · SMCF
(37) Cpro = Cep + Cprosp
(38) Mpro = Cpro · Qpro · SMCF
(39) Cep = Cp / (1 – pec)
(40) Mep = Cep · Qep · SMCF
(41) Cswep = Cep · ETCF
(42) Mswep = Mep
Initial soil water corrected by gypsum solubility
(43) Gypsum = A · Dr · Db · (PG)
(44) Mgyp = 2630 · Qpsw · SMCF
(45) If Mgyp
Gypsum : EC gype = 2.2 and TDSgyp = 2630 (Saturation)
(46) Cse = (ECe – EC gype) · 720
(47) Cisw = Cse · (SP · Dr · A · Db / Qisw)
(48) If Cisw < Cmin: Cisw = Cmin
(49) Misw = Cisw · Q isw · SMCF
(50) Cswisw = (Mswisw / Qswisw) / SMCF = Cisw · ETCF
(51) Mswisw = Misw
Gypsum, salt pickup-salt deposition
(52) Csp – Csd = a + b · LF
(53) Msp – Msd = (Csp – Csd) · Qeaiw · SMCF
(54) If Mgyp
Gypsum
Mgsp = Mgyp
Table 3 (cont). Mathematical equations describing the salinity submodel
Salt input to root zone
(55) Msw = Meaiw + Mep + Misw
(56) Csw = (Msw / Qsw) / SMCF
(57) Mpsw = Msw = Msweaiw + Mswep + Mswisw
(58) Cpsw = (Mpsw / Qpsw) / SMCF= Csw · ETCF
(59) M’psw = Mpsw + Mgsp + (Msp – Msd)
(60) C’psw = (M’psw / Qpsw) / SMCF
Final Soil water
(61) Cmin = ((Ceaiw + Csp - Csd) · Qeaiw + Cep · Qep ) / (Qsweaiw + Qswep)
(62) C’fsw = [k / (Dl / Dr) + k)] · (Cisw - Cmin) + Cmin
(63) NPV = (Dl / Dr) / [1 – (Db / 2.65)]
(64) If k = 0 and NPV < 1: C’fsw = [(1 – NPV) · Misw / Qfsw] / SMCF + Cmin
(65) Cfsw = C’fsw + TDSgyp
(66) Mfsw = Cfsw · Qfsw · SMCF
Salt output from root zone
(67) Mppsw = M’psw - Mfsw
(68) Cppsw = (Mppsw / Qppsw) / SMCF
(69) Cdp = Cppsw
(70) Mdp = Cdp · Qdp · SMCF
(71) Csdw = Cppsw
(72) Msdw = Csdw · Qsdw · SMCF
Rim inflow-outflows
(73) Mrim = Crim · Qrim
Salt load in irrigation return Flows
(74) Msirf = Miwro + Mpro + Msdw + Mrim
(75) Csirf = (Msirf / Qsirf)/ SMCF
For a value of k = 0, the model assumes a piston-flow displacement of salts from the
soil. Thus, the amount of salts leached from the soil depends of the number of pore volumes
(NPV) displaced with the water available for subsurface drainage and deep percolation (Qppsw).
To calculate NPV, a density of solids of 2.65 g/cm3 is used (Eq. 63). If NPV = 1, all the soluble
salts initially present in the soil (Misw) are displaced from the soil. If NPV < 1, the final
concentration of soil water (C’fsw) is composed of two terms: the first, proportional to (1-NPV),
gives the amount of salts initially present (Misw) that remain in the soil, and the second (Cmin)
gives the concentration of soil water in equilibrium with irrigation and precipitation waters (Eq.
64).
In the “excess gypsum” scenario, C’fsw does not include gypsum as Cisw and Misw have
been corrected by gypsum dissolution. The gypsum contribution to the saline concentration of
the final soil water is added through TDSgyp (TDSgyp = 2630 mg/L). For the rest of cases without
“excess gypsum”, it is assumed that TDSgyp = 0. The salt mass in the final stored soil water
(Mfsw) is given by Eq. 66.
The model does not consider a leaching efficiency coefficient for gypsum in the “excess
gypsum” scenario because it is assumed that drainage waters are gypsum-saturated. This is a
reasonable hypothesis supported by the ionic concentrations measured in waters draining from
soils high in gypsum (Basso, 1994, Quílez et al., 1987b, Isidoro et al, 2006).
The salt mass in the soil available for subsurface drainage and deep percolation (Mppsw)
is calculated from the difference between M’psw and Mfsw (Eq. 67) and its concentration (Cppsw)
from the ratio between Mppsw and Qppsw times the salt mass conversion factor (Eq. 68).
Once the mass of salts in the final stored soil water (Mfsw) is calculated by the model,
the soil solution is subdivided into two components of equal salt concentration (Cppsw): deep
percolation (Cdp), and collected subsurface drainage water (Csdw). The mass of salts in deep
percolation (Mdp) and subsurface drainage (Msdw) are obtained as the product of the above
concentrations times their respective volumes (Qdp, Qsdw) and the unit conversion factor (SMCF)
(Eqs. 70 and 72).
Salt load in surface irrigation return flows (Msirf) is the sum of the salt mass in
subsurface drainage (Msdw), runoff components (Mpro and Miwro), and lateral contributions (Mrim)
(Eq. 74) and the salt concentration of surface irrigation return flow (Csirf) is the volume-weighted
average of the concentrations in the three components (Eq. 75).
Table 4. Symbols used in the salinity submodel
Symbol
Definition
a
Intercept of the equation Csp-Csd= a + b · LF
b
Slope of the equation Csp-Csd= a + b · LF
Cdiw
TDS of diverted irrigation water
Cdp
TDS of deep percolation
Ceaiw
TDS of effective applied irrigation water
Cep
TDS of effective precipitation
Cfsw
TDS of final stored soil water
C´fsw
TDS of final stored soil water not including gypsum
Cisw
TDS of initial stored soil water, corrected for gypsum solubility if “excess gypsum”
Ciwro
TDS of irrigation runoff
Ciwrosp
Salt pickup by irrigation runoff
Cmin
Soil water concentration in equilibrium with irrigation and precipitation
concentrations
Cp
TDS of precipitation
Cppsw
TDS of soil water available for subsurface drainage and deep percolation
Cpro
TDS of precipitation runoff
Cprosp
Salt pickup by precipitation runoff
Cpsw
TDS of soil water after ET
C’psw
TDS of soil water after ET, Mgsp and Msp - Msd
Crim
TDS of rim inflow/outflow
Csdw
TDS of subsurface drainage
Cse
TDS of soil saturation extract corrected by gypsum solubility
Csirf
TDS of surface irrigation return flow
Table 4 (cont.). Symbols used in the salinity submodel
Symbol
Definition
Csp – Csd
Salt pickup – salt deposition
Csw
TDS of soil water before ET
Csweaiw
TDS of applied irrigation water after ET
Cswep
TDS of precipitation after ET
Cswisw
TDS of initial stored soil water after ET
Db
Soil bulk density
ECe
EC (25ºC) of soil saturation extract
ECgyp
EC (25ºC) of initial soil water due to gypsum if “excess gypsum”
ECgype
EC (25ºC) of soil saturation extract due to gypsum if “excess gypsum”
Gypsum
Amount of gypsum in the soil
k
Leaching efficiency coefficient
Mdiw
Salt mass in diverted irrigation water
Mdp
Salt mass in deep percolation
Meaiw
Salt mass in effective applied irrigation water
Mep
Salt mass in effective precipitation
Mfsw
Salt mass in final stored soil water
Mgsp
Amount of gypsum dissolved in soil water after ET
Mgyp
Amount of gypsum potentially soluble in soil water after ET
Mi
Salt inputs
Misw
Salt mass in initial stored soil water, corrected by gypsum solubility if “excess gypsum”
Miwro
Salt mass in irrigation runoff
Mo
Salt outputs
Mp
Salt mass in precipitation
Mpro
Salt mass in precipitation runoff
Mppsw
Salt mass available for subsurface drainage and deep percolation
Mpsw
Salt mass in soil water after ET
M'psw
Salt mass in soil water after ET, Mgsp and Msp – Msd
Mrim
Salt in rim inflow/outflow
Ms
Change in salt mass stored in the crop’s root zone
Msdw
Salt mass in subsurface drainage water
Msirf
Salt mass in surface irrigation return flow
Msp – Msd
Salt pickup - salt deposition
Msw
Salt mass in soil water before ET
Msweaiw
Salt mass in effective applied irrigation water after ET
Mswep
Salt mass in effective precipitation after ET
Mswisw
Salt mass in initial soil water after ET
Mua
Salt mass in IRF per unit irrigated area
NPV
Number of pore volumes
PG
Percentage of gypsum in the soil
SMCF
Salt mass unit conversion factor
TDSgyp
TDS of initial and final soil water due to gypsum if “excess gypsum”
3. INSTALLING CIRFLE
CIRFLE works under the Microsoft .NET framework, which only operates under the newer
Windows operating systems. Minimum system requirements are:
- Supported Operating Systems: Windows Server 2003; Windows Server 2008; Windows
Vista; Windows XP
- Processor: 400 MHz Pentium processor or equivalent (Minimum); 1GHz Pentium
processor or equivalent (Recommended)
- RAM: 96 MB (Minimum); 256 MB (Recommended)
- Hard Disk: Up to 500 MB of available space may be required
- CD or DVD Drive: Not required
- Display: 800x600, 256 colors (minimum); 1024x768 high color, 32-bit (recommended)
Insert the CIRFLE CD in the computer, double click on the file InstallCirfle.exe to initiate
the installation process, and then follow the on-screen directions. The installation program will
install CIRFLE in the ”program files” directory, but this directory can be changed during the
installation process.
CIRFLE needs the Microsoft Data Access (MDA) components and the .NET framework
be installed in the computer. The installation program will install the Microsoft data access 2.7 if
necessary. You have to accept the terms of the License agreement of the program by clicking in
the box “I accept all the terms of the preceding license agreement“ and then clicking on “Next”.
Once the setup program has checked the file used in the computer, click “Finish” to begin the
installation of MDA. When setup is completed, press “Close” to exit the MDA setup and continue
with the CIRFLE installation.
If the .NET framework is not installed in your computer a window with the message
“.NET runtime library is not installed¨” will open, click “OK” to continue. In the “Welcome to
Microsoft .Net Framework 2.0 setup” window click “Next” to start the Microsoft .NET Framework
2.0 installation. You have to accept the terms of the License agreement of .NET by clicking in
the box “I accept the terms of the License agreement“. Then click “Install”. This will install the
components of the .Net framework in your computer. Please be patient, the installation of .NET
Framework can take time and resources used in your computer. When the “Setup Complete”
window appears in your screen press “Finish” to continue with the CIRFLE setup.
When the setup program finishes installing CIRFLE you should click “Finish” to exit the
program. Double click in the shortcut in the screen desktop to start using CIRFLE.
Note that if you have other background programs running during the installation
process, CIRFLE may not install properly. If you encounter an error message "file access error
occurred", please close all the background programs running in you computer.
4. RUNNING CIRFLE
4.1. CIRFLE application
The first step is to identify the time interval or simulation period for which the model is to be
applied. CIRFLE has been designed to be applied to large systems and for long periods, such
as an irrigation season, a hydrologic year, or a series of consecutive years. The model cannot
be applied for short periods as, in general, steady state conditions do not hold.
The model's variables and parameters are spatial and temporal averages. Thus, CIRFLE
must be applied to areas with homogeneous characteristics or the results can be misleading.
So, the second step in the application of CIRFLE is the identification and delimitation of areas
with similar characteristics within the study area. Special attention should be paid to (1) soil
characteristics (in particular, ECe, presence or absence of native gypsum, salt leaching
efficiency, k, and soil depth, Dr); (2) method of irrigation (surface, sprinkler, drip, furrow,..) that
affects the volume of applied irrigation water, the irrigation application efficiency or the irrigation
evaporation coefficient; and (3) crop characteristics that determine the volume of
evapotranspiration and the rooting depth.
Based on this and other variables, the irrigation project should be divided in as many
homogeneous areas as necessary. Inputs for each area must be entered separately in the
“Input Data for the Area” screen. After calculation, CIRFLE returns the outputs in the “Output
Data for the Area” screen. Once the volume of water, salt concentration and mass of salts in the
surface irrigation return flows of each area are computed, CIRFLE integrates the values of all
the areas to give the volume of water, salt concentration and salt load in the surface irrigation
return flows for the whole project in the “Irrigation Project Output” screen.
After an introductory window, you will find the main screen of the program. Two menus
are included in the upper bar of the program: “Project” and “Area”.
The Project menu has six options:
- “New project”, to create a new project.
- “Open project”, to retrieve a saved project.
- “Save project”, to save data for a new project or to save modifications to already
saved projects.
- “Delete project”, to delete a project
- “Generate output project”, to save the output of the project to a pdf or excel file.
- “Exit”, to exit the program
The main screen has two sections. In the left part of the screen, you will see the name
of the Irrigation Project that unfolds in as many areas as the project is sub-divided; each of the
areas is identified by its name. When you select in the left screen one of the Areas, in the right
hand side of the screen you activate two tabs. The first tab “Input Data for the Area” shows the
25 variables or parameters that are inputs to the model. The second tab “Output data for the
Area” gives the outputs of the model for the selected Area.
When you select the name of the Project in the left part of the screen, in the right part
you get also two tabs. The “Irrigation Project Input” presents the different Areas in which the
project has been divided, with information of its surface area and the volume of water applied for
irrigation. The “Irrigation project Output” gives the volume, saline concentration and mass of
salts in the irrigation return flows individually for each of the Areas and integrated for the entire
Irrigation Project.
4.2. “Project” menu
New project
The “New Project” option assists to input data for the new projects. When you press the
new project tab a window opens asking for some basic information on the project that should be
answered (Figure 3). First, you have to write the name of the project in the space designated in
the blank line under New Project. Use only letters and numbers for the name of the project, the
use of special symbols like “/” can produce errors. Then, you have to choose the units of the
input data. It is possible to choose between four different units for water volume: Hm3, mm,
m3/ha and ha-m. The unit ha-m (1 hm3 = 100 ha-m) is preferred over hm3 when the hm3 unit is
too large for the size of the irrigation project. You select the unit by clicking with the mouse in
the circle located on the left of the unit you want to select. You will have to enter all the water
volume data (Q) in that unit. In addition, all the outputs for volume of water will be given in the
same unit.
Figure 3. “New project” menu
Although CIRFLE works with salt concentrations (C) in mg/L, there are two possibilities
to enter salinity: salt concentration in mg/L and electrical conductivity in dS/m. To choose a
particular unit, click with the mouse on the white circle on the left of the desired unit. If you
choose EC units for the saline concentration of the input variables (Cdiw, Crim, Ciwrp, Cp, Cprosp)
you have to enter the EC values in dS/m. You have also to enter a conversion factor (Fc) to
convert EC into C [C (mg/L) = Fc·C (dS/m)]. Conversion factors for seven different types of
water ionic compositions are included in CIRFLE (Tabla 10).
Finally, the number of homogeneous areas that constitute the project need to be chosen
by clicking in the down arrow located at the right of the cell “number of homogeneous areas in
the project area” and choosing a number in the opening list. Once you choose the number of
areas the tag “Create project” will be active. Click
Create Project
and a new window will open.
In this window you can introduce the names of the different areas of the project (as many as you
entered in the previous step). Once you are finished, click
Create Areas
at the bottom of the
screen to start entering the data to the project. If you do not see Create Areas enlarge the
window downwards. Use only letters and numbers for the name of the area, the use of special
symbols like “/” can produce errors.
The name of the project will now appear in the left window, and below the different
Areas that integrate the project will unfold. The input data window for each of the areas will open
in the right window by clicking the name of the area in the left window. In the upper part of the
“Input Data for the Area” window, the name of the area appears over the input frame that
encloses the 25 input variables. In addition, the conversion factors (fc) from EC to TDS for
variables Cdiw, Cp and Crim should be input when the variable selected for salt concentration be
EC. You will find in section 4.4. “Input Data For the Area” how to enter the data in this window.
Open project
With the open project option, you can retrieve projects already saved. Projects are
stored in a relational data base file (cirfledb.mdb) that contains the name of each Project and
the input data for the Areas of each project. When you choose the open project button, you get
a list with the names of the projects stored in the database. To select one project click in its
name and then in
Open Project
; the data for that project will be retrieved in the different
windows. You can modify the values of the input variables in any of the Areas, or add new
Areas to the Project. When you modify the value of a variable be sure to press Calculate to get
the modified outputs in the “Output Data for the Area” tab. If you do not press the Calculate
button, the outputs will not be modified.
Save project
This tab saves the data of the project in the database file, i.e., the name of the project
and the input data for each of the areas. If you want to modify some of the input data for a
project already saved you should open the project, change the value of the variable(s) you want
to modify, press Calculate , and press the save project tab to store the modified inputs.
Generate output project
This tab saves the output of the project in a pdf or xls file The pdf file present in the first
page a resume of the volume, salt concentration and salt load in irrigation return flows from the
different areas that integrate the project and in the next pages the detailed outputs of all the
intermediate variables and the graphic outputs for each of the areas of the project. In the xls
format the detailed outputs are presented in different sheets, one for each area of the project.
The “generate output project” can also be found in the lower part of the “irrigation project output”
window when selecting the whole project on the left part.
4.3. “Area” menu
Using the “Area” menu, you can create a new area or delete an area in your project. If you
select the tab “Create Area”, a window opens where you should enter the name of the new area
and then press
New Area
to create the new Area. Selecting the new Area in the left of the
screen you can input its corresponding data in the right part of the screen (Figure 4). To delete
an Area just select the name of that area in the left part of the screen and then the option
“Delete Area” under the “Area” menu.
4.4. “Input Data For the Area” menu
The “Input Data for the Area” (Figure 4) is active when you select one of the Areas in
the left of the main screen. This window allows to enter the values of the input variables for an
Area. If the user opens a Project it will show the input data for the areas defined in the Project.
In the user creates a new project, all the cells for the values of the input variables will be blank.
The “Input Data for the Area” window presents the name of the Area in the first line.
Below the name, you will find the 25 inputs to the model. By clicking in the cell close to each
input, a description of it is given below “description”. At the bottom of the screen, there are three
buttons: Calculate that computes the outputs for the model, Sensitivity Analysis that performs
the sensitivity analysis of the model, and
New Water Type
that allows the user to define new
conversion factors to transforms EC into TDS.
You have to enter the values of each input in the blank cells located in the right side of
each input. Use always “.” for the decimal symbol. The Inputs are given by acronyms, but a full
description of each one is given by clicking inside each cell. This description helps to identify the
acronyms used for the names of the inputs. At the right of each input cell, you find the units for
each of the input variables, except for the dimensionless ones (iwec, prc, pec, k); be sure to
enter the value of each variable in the units indicated in the screen.
Figure 4. “Input data for the Area” menu
If you selected the option EC for saline concentrations (section 4.2), it is necessary to
introduce a conversion factor (fc) to transform EC into TDS for each of the five input
concentrations (C). For each C, you have to select one conversion factor in the unfoldable menu
at the right of that concentration. CIRFLE includes seven most frequent types of water (Table
10). If you know the fc for your water type, you can input it by clicking the New Water Type .
button located at the bottom right corner of the screen.
The parameters a and b of the equation Csp – Csd = a + b LF are located in the middle of
the right column. When you press inside the input cell of anyone of these two parameters, you
will see in the description line the values of a and b estimated for different irrigation water types
(Table 9). You can use this values or introduce you own values.
4.5. “Output Data for the Area” menu
The “Output Data for the Area” screen presents the outputs for each of the Areas of a
project. It is only active when you have previously selected an Area in the left part of the screen.
The outputs for the model are given in four columns (Figure. 5). The first column gives the
volume of water Q (in the units selected for the project), the second the saline concentrations C
(mg/L), the third the mass of salts M (tons), and the fourth some parameters of interest. A
description of each parameter is given when clicking in each of them.
Figure 5. “Output data for the Area” menu
The Graphic
button in the lower right prompts a graphical summary of the water and
salt fluxes in the Area resulting from the simulation. The graphic output is a diagram that
represents the flows of water and salt in the system following the scheme presented in Figures
2 and 3. To save the outputs from an area you need to save the whole project in the “Irrigation
Project Output” tab (see 4.7)
4.6. “Sensitivity Analysis” menu
The Sensitivity analysis gives the option to analyze how variations in the values of the
input parameters will affect the outputs of the model for a selected Area.
To perform a sensitivity analysis (SA) press the
Sensitivity Analysis button located at
the middle bottom of the Input data for Area screen, and the SA window will open. Under the
Options menu located in the lower left of the screen you can select either a manual or an
automatic SA. In both cases you can select up to 2 variables to modify in the two boxes under
the select variables line. The variables that can be changed are Qdiw, Eiae, Qp, prc, pec, Qisw,
Qfsw, Dr, Qet, dpc, Qrim, Crim, Cdiw, Ciwrosp, Cp, Cprosp, ECe, SP, Db, Gypsum and k.
Under the manual option, the boxes corresponding to the variables selected will activate
so that you can enter the new value(s) of the variable(s). In the automatic option, you have to
define an interval to perform the SA entering a lower limit and upper limit and the interval.
If you select only one variable, the automatic SA will modify the value of the variable
from a minimum set value (lower limit * value of the variable /100) to a maximum set value
(upper limit * value of the variable /100) using fixed interval increments (interval * value of the
variable /100). The number of different values obtained (N) is defined by N = ((UpperLimit –
Lower Limit) / Interval)). The SA will calculate the outputs for the resulting N values of the
variable. If you select two variables, the same values of lower limit, upper limit and interval are
used for the two variables. The SA will calculate the outputs for the resulting N2 combinations of
the values of the two variables.
When you press
Calculate
in the bottom center, a window will open asking for the
format in which you want to save the output (excel or pdf). You should select one, as well as the
folder where you want to save the resulting output. The output file will be stored in the selected
folder with the name of the AREA.
The excel file presents a complete list of the inputs,
intermediate, and outputs variables, and the management parameters for the different values of
the variable(s). The pdf file presents first a summary output of the volume, salt concentration
and salt load in irrigation return flows (Qsirf, Csirf, Msirf, respectively) for the different values of the
variable(s) and then the completed output.
4.7. “Irrigation Project Input” and “Irrigation Project Ouput” menus
When you select the name of the Project in the left side of the main window, two tabs
are activated in the upper part of the window: “Irrigation Project Input” and “Irrigation Project
Output”.
The “Irrigation Project Input” menu records the different Areas in which the project has
been divided, with information of the surface area (Area) and the volume of diverted irrigation
water (Qdiw) for each Area, as well as the surface area and the volume of diverted irrigation
water for the entire Irrigation Project .
The “Irrigation project Output” gives for each Area (Area Name) the volume (Qsirf), salt
concentration (Csirf), mass of salts (Msirf), mass of salts per unit irrigated area (Mua) and Mass
of salts per unit water inflow (mg/L) in the irrigation return flows for each of the Areas of the
Irrigation Project and for the entire Irrigation Project. Qsirf and Msirf for the entire Irrigation
Project is obtained by adding the volume and salt mass for the individual Areas, and Csirf is
obtained from Msirf/Qsirf (volume weighted average concentration).
You can save the outputs of the projects to a file selecting the Generate Output Project .
at the bottom of the screen. You can save the outputs in pdf or excel formats. The pdf file
present in the first page a resume of the volume, salt concentration and salt load in irrigation
return flows from the different areas that conform the project, and in the next pages the detailed
outputs of all the intermediate variables and the graphic outputs for each of the areas of the
project. In the xls format the detailed outputs are presented in different sheets, one for each
area of the project. The information is saved in a file with the name of the project. If that file
name already exist in the directory consecutive numbers (1,2 ….) area added at the end of the
project name.
5. DESCRIPTION OF INPUTS
CIRFLE requires 25 inputs variables and parameters for each of the homogenous Areas
in which the Irrigation Project is disaggregated. Figure 4 shows the “Input data for the Area”
menu and Table 5 summarizes the names of the variables and their acronyms. A description of
each of these inputs variables and parameters follows.
Table 5. Variable names and acronyms of CIRFLE input variables and parameters.
#
Variable name
Acronym
1
Irrigated surface
Area
2
Volume of diverted irrigation water
Qdiw
3
Irrigation water evaporation coefficient
evdiw
4
Irrigation application efficiency
Eiae
5
Volume of Precipitation
Qp
6
Precipitation surface runoff coefficient
prc
7
Precipitation evaporation coefficient
pec
8
Initial stored soil water
Qisw
9
Final stored soil water
Qfsw
10
Average crop’s rooting depth
Dr
11
Volume of total real crop’s evapotranspiration
Qet
12
Deep percolation coefficient
dpc
13
Volume of surface rim inflows and outflows
Qrim
14
Saturation Percentage
SP
15
Soil bulk density
Db
16
Gypsum percentage in the soil
Gypsum
17
Salt leaching efficiency coefficient
k
18
Intercept of the equation Csp - Csd= a + b · LF
a (Csp-Csd)
19
Slope of the equation Csp - Csd= a + b · LF
b (Csp-Csd)
20
Salt concentration (or EC) of irrigation water
Cdiw
21
Salt concentration (or EC) of surface rim inflows
Crim
22
Salt pickup by irrigation runoff
Ciwrosp
23
Salt concentration (or EC) of precipitation
Cp
24
Salt pickup by precipitation runoff
Cprosp
25
Electrical Conductivity of soil saturation extract
ECe
5.1. Inputs defining the water balance
Numbers before each input refer to # in Table 5. Remember to use always “.” for the
decimal symbol.
1- Irrigated surface (Area)
Refers to the surface area of the irrigation district, that is the area that can be irrigated
during the period of simulation. Always introduce the Area in hectares (ha).
2- Volume of diverted irrigation water (Qdiw)
Refers to the total volume of water diverted for irrigation in the simulation period
(normally yearly or seasonal volume). It is given in the same units chosen in the “New Project”
menu [hm3, mm, m3/ha or ha-m (hectare · m)]. Note that the units hm3 (1 hm3 = 106 m3) and ham (1 ha-m = 104 m3) are units of volume, whereas mm (1 mm = 10 m3/ha) and m3/ha are units
of volume per unit area.
The volume of water applied through each of the different water delivering systems
should be measured and added to obtain the total volume of diverted irrigation water in each
area during the studied period.
3- Irrigation water evaporation coefficient (evdiw)
Refers to the fraction (0 to 1) of Qdiw that is lost as winds drift or that evaporates form
the crop canopy, soil or the distribution systems before it infiltrates the soil.
Although evdiw is generally low in surface and drip-irrigated systems, it may be high in
sprinklers systems subject to high winds. Thus, wind drift and evaporation losses (WDEL) may
be as high as 0.3 (Faci and Bercero, 1989) to 0.4 (Dechmi et al. 2004). In order to estimate
WDEL in high-wind areas from the average wind velocity, the following regression (Dechmi et
al., (2004) may be used: WDEL = 12.02 W 0.626 [R2 = 0.79], where W is the wind speed at 2m
height. This equation has been obtained at the plot scale; however, at the irrigated district scale
(normally the usual size for CIRFLE application) these losses could be much lower as the
losses in one plot may be gains in the neighboring plots. Whether WDEL are actual losses or
not depends mainly on drop size (smaller drops tend to evaporate and larger ones tend to fall).
4- Irrigation application efficiency (Eiae)
Refers to the fraction (0 to 1) of Qdiw that actually infiltrates the soil. Besides wind drift
and evaporation losses already mentioned in evdiw, Eiae must account for all losses in the
distribution system after reading of the water meters (OL), such as secondary canal seepage or
operational losses, tail-waters from irrigation ditches (TW) and irrigation runoff at the plot scale
(Runoff). The volume fraction of all these losses must be subtracted from one to obtain Eiae as:
Eiae = 1 – (TW + OL + Runoff) – evdiw). If the sum of Eiae and evdiw is more than 1, CIRFLE
prompts an error message.
The value of Eiae should be established for each Area. If actual measurements are not
available you can use sensible guesses of these losses. A few directions on their magnitude for
different irrigation systems follow:
(i) Flood irrigation systems with open channel distribution: the losses take place both as
operational losses (OL) at the end of the irrigation ditches and as runoff from irrigated
plots —tail-waters (TW). The sum of both flows (and possibly some seepage from the
main canals) was found to be 17% of Qdiw in a fixed-schedule distribution system in
Spain (Isidoro et al., 2004) [Thus: Eiae = 0.83 – evdiw]. Lecina et al. (2004) found that
surface runoff in flood irrigated basins in the Bardenas district (Spain) was about 10.6%
in coarse-textured soils with high infiltration rates, and 11.2% in fine-textured valley soils
[Thus: Eiae = 1 – 0.106 (or 0.112) – OL – evdiw]. In systems with fixed irrigation shifts
(where farmers receive water on fixed dates and must use it whether or not they actually
need it on that date), OL is likely higher than in more flexible systems. Normally, higher
irrigation flows will lead to reduced percolation losses and enhanced uniformity, but will
increase TW. In border- or basin-irrigated plots (with no outlets at the end of the plots),
TW = 0. In furrow irrigation systems (with opened ends), siltation in the furrow bottoms
may enhance runoff (Al-Qinna and Abu-Awwad, 1998).
(ii) Sprinkler irrigation systems: OL should be negligible whereas the magnitude of TW
depends on the slope of the fields, soil’s infiltration rate and the right application of the
irrigation. When the irrigation rate or the sprinkler discharge (mm/h) is lower than the
soil’s infiltration rate (i.e., adequate design of the irrigation system), TW should also be
negligible. But in other situations TW may be high. Schneider and Howell (2000) found a
mean runoff of 12% in 0.25% slope furrows on a clay loam soil (Texas, USA) and 22%
runoff in low-energy precision application sprinkler systems (when irrigation was 100% of
crop needs). Ben-Hur et al. (1995) documented runoffs varying from 0 in the first
irrigation to 37.5% in the later ones in a 3% slope silt loam soil in Neguev (Israel). As
crust formation leads to higher runoff, in areas where crust formation is known to be a
problem, Eiae should be reduced accordingly (higher TW). Al-Qinna and Abu-Awwad
(1998) reported up to 20.3% runoff values in a sprinkler irrigated experiment with a low
sprinkler rate of 6.2 mm/h and up to 48.3% with a high sprinkler rate of 28.4 mm/h rate in
fine silt soils with surface crust problems in Jordan.
(iii) Drip irrigation systems: OL, TW and evdiw should be negligible or very low.
5- Volume of precipitation (Qp)
Refers to the volume of precipitation during the simulation period expressed in the
selected units. Qp should be measured in different locations distributed in the study area since it
may vary depending on climatic and topographic characteristics.
6- Precipitation surface runoff coefficient (prc)
Refers to the fraction (0 to 1) of Qp that becomes runoff over the soil surface, without
infiltrating the soil. This term must account for the total runoff in the Area during the simulated
period; it may be estimated from surface runoff estimation methods such as the Curve Number
method.
7- Precipitation evaporation coefficient (pec)
Refers to the fraction (0 to 1) of Qp that evaporates before infiltrating the soil. Both prc
and pec define the proportion of Qp that actually infiltrates the soil [(1 – prc – pec)]. When the
input data for prc and pec are such that prc + pec > 1, CIRFLE prompts an error message.
8, 9- Volume of initial (Qisw) and final (Qfsw) stored soil water
They refer to the mean root-zone soil water contents at the beginning (Qisw) and end
(Qfsw) of the simulation period. The unit is percent weight or gravimetric water content
(g H2O/100 g dry soil or cm3 H2O/100 g dry soil).
When the simulation period is the hydrological year (the recommended period for
CIRFLE application), Qisw and Qfsw can be taken as equal (assuming that the water contents in
the same date of consecutive years are approximately equal). For long-term simulations of
average conditions, it may also be assumed that Qisw = Qfsw.
When the simulation is performed for periods lower than a hydrologic year (it is not
recommended to use CIRFLE for periods shorter than an irrigation season ~ 6 month) an
estimate or actual field measurement of both Qisw and Qfsw is needed. Determining Qisw and Qfsw
in large irrigated areas is costly and time consuming. The more homogeneous the selected
area, the lower the number of samples needed. In absence of actual soil water measurements
and for a hydrological year period, using field capacity (FC) for both Qisw and Qfsw is
recommended. In all cases, the range of values should be normally above the wilting point (WP)
and below field capacity (FC) of the soil.
As a help for the selection of Qisw and Qfsw when no measured data are available and
only the general soil properties are known, Table 6 includes the usual range of WP and FC for
different soil textures. A rule of thumb if other information is not available is that FC is about two
times WP.
10- Average crop’s rooting depth (Dr)
Refers to the mean rooting depth (m) of the crops grown in the irrigated soils of the
selected Area. For very deep soils, Dr is taken as the average rooting depth of the crops. For
shallow soils limiting root growth, Dr is taken as the average depth of the growth-limiting soil.
Table 6. Usual ranges (minimum, maximum and mean) of gravimetric water contents at
field capacity (FC) and permanent wilting point (WP) for several soil textures (Source:
FAO, 1979)
FC (cm3 H2O/100 g dry soil)
Texture Minimum Average Maximum
WP (cm3 H2O/100 g dry soil)
Minimum Average
Maximum
Sandy
6
9
12
2
4
6
Sandy Loam
10
14
18
4
6
8
Loam
18
22
26
8
10
12
Clay Loam
23
27
31
11
13
15
Silty Clay
27
31
35
13
15
17
Clay
31
35
39
15
17
19
11- Volume of total real crop’s evapotranspiration (Qet)
Refers to the volume of water actually evapotranspired by the irrigated crops. It includes
the evapotranspiration (ET) of all the irrigated crops in the selected Area and the ET of natural
vegetation (if it is known and deemed important). If there are several crops in the Area, Qet is
the sum of the ET’s calculated for each crop (when given in hm3/ha or ha-m) or the areaweighted Qet for each crop (when given in mm or m3/ha).
It should be emphasized that Qet refers to the actual or real ET, not ETo (reference ET)
or ETc (maximum or potential crop’s ET calculated from ETo and the crop’s coefficients, Kc).
Note that this term is critical in defining the water balance in the irrigated soils of the Area and it
must be estimated as accurately as possible.
We suggest following the FAO guidelines (Allen et al, 1998) to calculate the ETc for
each crop. In Areas where crops are not subject to any kind of stress (in general, an unrealistic
scenario), it may be accepted that the real crop ET (Qet) is equal to ETc. This could be
approximately the case for drip-irrigated or sprinkler-irrigated areas with frequent irrigations that
avoid water stress. In areas where irrigation intervals are longer and crops suffer of water
stress, Qet is lower that ETc and the lower values should be estimated according to the
observed lower yields. In many instances a soil bucket-type balance in the soil (based on
estimates of FC and WP and the actual distribution of rainfall and irrigation for each crop; Allen
et al, 1998) may be sufficient to estimate Qet. Other biotic or abiotic stresses that may reduce
yield and ETc should also be taken into account if possible, to obtain a better estimate of Qet.
12- Deep percolation coefficient (dpc)
Refers to the fraction of deep percolation waters (waters flowing below the crop’s root
zone) that is not collected by the surface drainage systems present in the study areas and, thus,
do not contribute to to the surface return flows. This “percolating” volume flows towards deeper
regional aquifers not linked to the aquifers intercepted by the drainage system.
For irrigation systems underlain by impervious materials or in the absence of important
and conductive groundwater systems, it can be assumed that dpc = 0. If during CIRFLE
calibration the error in the water balance is high and the predicted outputs (Qsirf) are noticeably
lower than the measured outflows, it might well happen that some fraction of the deep
percolation waters are not intercepted by the drainage system. In these cases, one step of the
calibration procedure may be to set a dpc value that will decrease the error in the water balance
(i.e. dpc > 0).
13- Volume of surface rim inflows and outflows (Qrim)
Refers to lateral surface inflows into the study area or lateral shallow ground-water
inflows that are collected by the surface drainage system therefore contributing to the outflows
from the irrigated area.
In the calibration process of CIRFLE, large errors in the water balance (i.e., Qsirf higher
than measured outflows) may be potentially explained by the presence of significant
unmonitored lateral flows into the Area. Surface rim inflows can be monitored by means of
gauging stations, but shallow groundwater inflows are difficult to estimate. In dry climates and
when the contributing dry land area is small, groundwater Qrim could be neglected.
Often, in old irrigation systems, especially with unlined canals, seepage may be
substantial and it should be inputted in CIRFLE as Qrim. If the rest of the water balance
components are well known, Qrim may be estimated as the error in the water balance. Lateral
inflows from outside the area should include an estimate of the seepage from the main canals
(upstream of the diversions where Qdiw is actually measured). This seepage can be very
important in old irrigation schemes and especially in systems with earth (unlined) conveyance
structures.
5.2. Inputs defining salt loading and salt leaching
14- Saturation percentage (SP)
Refers to the amount of distilled water added to an air-dry, ground and sieved (< 2 mm)
soil to prepare the saturated paste (units of g of water per 100 g of dry soil).
If the SP is not determined in the lab, it may be estimated from the Sand fraction (0 to 1)
in the soil through the regression: SP = 1.03 - 0.796 · Sand [R2 = 0.63] obtained from the data of
Banin and Amiel (1969) for 33 soil samples in Israel. These authors reported SP’s in the range
from 25% (soil with 98% Sand) to 114% (soil with 67% Clay). If the texture of the soil is known,
the SP range for each soil textural class may be estimated from Table 7.
Table 7. Usual ranges of SP for several soil textural classes (source: Slavich and
Petterson, 1993).
Textural class
SP range (%)
Sandy, Loamy sand, Clayey sand
<20
Sandy loam, Fine sandy loam, Light sandy clay loam
20 - 41
Loam, Loam fine sandy, Silt loam, Sandy clay loam
41 - 46
Clay loam. Silty clay loam, Fine sandy clay loam, Sandy clay, Silty clay,
Light clay, Light medium clay
46 - 53
Medium clay
53 - 72
Heavy clay
> 72
15- Soil bulk density (Db)
Refers to the average bulk density of the soil for the whole rooting depth (units of g/cm3).
Db is required to convert Qisw and Qfsw (% weight) into volumes of water and to determine the
amount of gypsum (in Mg) present in the soil from the % Gypsum given as an input.
Table 8 (National Resources Conservation Service, NRCS-USDA) gives some
guidelines to estimate bulk density from textural classes when Db is not measured. As indicated
by NRCS the guidelines provide estimates for moderately consolidated soils (i.e. moderate
grade of structure and generally friable consistence).
Table 8. Usual ranges of Db for several soil textural classes (source: Natural Resources
Conservation Service, USDA.
http://www.mo10.nrcs.usda.gov/references/ guides/properties/moistbulkdensity.html)
Texture
coarse sand
sand/ fine sand /
loamy coarse sand
Bulk density
1.70 - 1.80
1.6 - 1.7
1.55 - 1.65
sandy loam /
fine sandy loam
very fine sandy loam /
loam / silt loam / sandy
clay loam /silty clay loam
Texture
loamy very fine sand
coarse sandy loam
very fine sand / loamy
sand / loamy fine sand
Bulk density
1.55 - 1.6
1.5 - 1.6
1.45 - 1.55
Texture
silt / clay loam /
silty clay
sandy clay /
clay (35-50%)
clay (35-50%)
Bulk density
1.40 - 1.5
1.35 - 1.45
1.25 - 1.35
These estimates will be higher in soils that have high or very high consolidation (i.e.
weak structure or massive and generally very firm consistence [e.g. "Cd horizon, natric
horizon"]) and lower in soils with strong structure and generally loose consistence.
16- Gypsum percentage in the soil (Gypsum)
Refers to a dichotomous input that accounts for the presence (Yes) or absence (No) of
gypsum in the soil. If the gypsum percentage (PG) is known, it can be inputted in the window to
the right of the Gypsum box. If the Yes button is activated but the % Gypsum is not introduced,
CIRFLE uses a default value of 20% to guarantee gypsum saturation. As the soil solution
becomes easily saturated with gypsum with low gypsum contents in the soil, this % is usually
not important and the actual value must not be specified. Only for very low gypsum contents in
the soil (gypsum < 1%) it should be specified.
Gypsum dissolution is an important contribution to the salinity of soil water. Thus, the
presence of gypsum in the soil must be assessed beforehand. In semi-arid environments, when
the 1:5 extract yields an EC of around 2 to 2.5 dS/m, the presence of gypsum is almost certain.
17- Salt leaching efficiency coefficient (k)
CIRFLE estimates the mass of salts leached from the root zone during the simulation
period using equations 61 to 68. The model uses the empirical approach developed by
Hoffman (1986) to estimate the final concentration of soil water (C’fsw) from the initial
concentration of soil water (Cisw) by using a leaching efficiency coefficient (k) and considering
the amount of water that percolated below the crop’s root zone (Qppsw/A) per unit depth of
average crop’s rooting depth (Dr) as:
C 'fsw - Cmin
k
=
Q
Cisw Cmin
ppsw
Dr
A +k
The leaching efficient coefficient is an empirical parameter that depends on soil factors
(structure, texture, salinity) and management factors (irrigation systems and management). As
k increases, the efficiency of leaching decreases. For k = 0, piston flow is simulated (i.e.,
leaching efficiency = 100%). For k =
, a hypothetical null displacement of salts is assumed
(i.e., leaching efficiency = 0%).
Typical k values for continuous ponding are around 0.1 for sandy loam soils (i.e., high
leaching efficiency), 0.3 for clay loam soils (i.e., moderate leaching efficiency) and 0.45 for peat
soils (i.e., low leaching efficiency). In intermittent ponding and sprinkler systems where
leaching takes place at soil water contents below saturation, k is around 0.1 (i.e., high leaching
efficiency) and independent of soil texture. If the value of k is too low (k<0.1); the estimated
concentration of subsurface drainage and deep percolation can be untruthful and extremely
high.
18, 19- Intercept (a) and slope (b) of the salt pick-up minus salt deposition vs. leaching
fraction (LF) equation [a (Csp-Csd), b (Csp-Csd)]
Up to this point, CIRFLE has only considered gypsum dissolution if present in the soil.
However, calcite is frequently present in many arid and semiarid soils and may be dissolved by
the percolating waters. Also, gypsum and calcite may precipitate in the soil depending on the
concentrations of Ca, HCO3 and SO4 in the irrigation waters and on the ETCF (concentration
factor of the irrigation water in the soil due to crop’s evapotranspiration). The potential
dissolution (“salt pick-up”, Csp) of the calcite present in the soil and the potential precipitation
(“salt deposition”, Csd) of calcite and gypsum in the soil due to the ETCF of the irrigation water
are accounted empirically through an equation that relates Csp-Csd with the leaching fraction:
(Csp-Csd) = a + b · LF.
Inputs to the CIRFLE model are the coefficients a (intercept) and b (slope) that are
included in the model for different types of water (Table 9). These coefficients were obtained
through simulations with the model Watsuit (Wu et al., 2009) for several water types present in
the Ebro River Basin and in many other arid and semiarid basins around the world. For each
water type, the mean irrigation water concentrations of Ca, Mg, Na, K, Cl, SO4 and HCO3 were
used as inputs and run with the option “saturated with calcite” that allows for the dissolution of
calcite up to its chemical saturation. Five LF values were used (0.05, 0.10, 0.20, 0.30, 0.40) to
obtain the linear regression equations (Csp-Csd) = a + b · LF. Table 9 gives the EC range, slope,
intercept and coefficient of determination values of these regressions.
These water types cover many usual irrigation waters, but there may be other types of
irrigation waters. If the exact composition of the irrigation water applied in the study area is
known, (Csp-Csd) may be estimated using Watsuit. The results for calcite dissolution and
precipitation and gypsum precipitation for the five soil boundaries in Watsuit must be weighted
by the leaching fraction computed by Watsuit for each boundary. Take into account that the sign
convention in Watsuit is the opposite to the sign convention in CIRFLE: deposition is negative
and pick-up is positive. Obviously, if a local experimental relationship between Csp-Csd and the
actual LF is known, it should be used instead of those indicated in Table 9.
In general, the contribution of Csp-Csd to the overall salt fluxes is low. Even so, the
values of a and b can be modified in the final, fine-tuning of the calibration process.
Table 9. Parameters (a and b) and coefficients of determination (R2) of the regressions
Csp-Csd = a + b · LF for each water type.
EC range (dS/m)
a
b
R2
HCO3-Ca
0.10 – 0.45
-107
366
99%
HCO3-SO4-Ca
0.45 – 0.70
-223
478
99%
HCO3-SO4-Cl-Ca-Na
0.70 – 1.05
-206
463
99%
SO4-HCO3-Cl-Ca-Na
1.00 – 1.40
-386
670
96%
SO4-Ca
1.30 – 1.90
-353
611
99%
Cl-HCO3-Na-Ca
1.00 – 1.40
-517
1263
89%
> 2.00
-2019
3932
99%
Water type
SO4-Ca (hyper SO4)
25- Electrical conductivity of soil saturation extract (ECe)
Refers to the concentration of soluble salts obtained in a soil saturation extract (units of
dS/m). CIRFLE convert ECe (dS/m) into TDS (mg/L) through the factor FC = 702 (mg/L) / (dS/m).
This Fc value was obtained as the mean for many waters with typical ionic compositions in the
Ebro river basin.
Soil salinity is generally highly variable in time and space, and may be the most important
salt source in many arid and semiarid study areas. Thus, the delineation of the homogeneous
areas is usually performed based on ECe. To this end, time and space-averaged ECe values for
each delimited area should be estimated with the maximum reliability. The spatial variability of
salinity in the study area should be analyzed in detail, using all the information available
(geomorphology, soil maps, etc.) and, if possible making surveys with electromagnetic devices,
the best way to obtain the needed soil salinity maps.
5.3. Inputs defining salinity in the input water fluxes
20, 21, 22, 23 and 24- Salt concentrations in irrigation, precipitation and rim inflows
Salt concentrations of the irrigation waters should be monitored for all the sources of
irrigation. The average salt concentration of irrigation water (Cdiw) is the volume-weighted
average of these salt concentrations. Salt concentrations in precipitation (Cp) and rim inflows
(Crim) must be determined in the same way.
The fraction of irrigation and precipitation flowing over the soil surface as surface runoff is
assumed to pick-up some salts as they convey towards the surface drainage system. These
increases in salt concentration (Ciwrsop for irrigation water runoff and Cprosp for precipitation
runoff) are typically low. The term Ciwrosp refers to the salt pick-up in runoff from irrigated plots,
operational losses and seepage from secondary irrigation ditches (after the water meters
defining Qdiw). In general, this pick-up term is low, compared to other salt sources. It can be
estimated measuring EC at the input and output points of irrigation in some fields.
5.4. Conversion factors for estimating TDS from EC
In the “New Project” menu (Figure 3), the user may choose between TDS or EC as
“Units for Saline concentration”. If EC is selected, in the “Input data for the Area” menu the salt
concentrations Cdiw, Crim, Cp, and are given in dS/m and they must be converted into TDS in
order to obtain salt loads using the Fc values given in Table 10. According to the ionic
compositions for each water, the appropriate Fc values are introduced in CIRFLE in the
windows to the right of each of the above salt concentrations. If FC is known by the user for the
irrigation water diverted to the study area, it can be introduced directly into CIRFLE with the
option “New Water Type” shown at the bottom right of the “Input Data for the Area” menu.
Table 10. Water types and conversion factors (FC) to convert electrical conductivity (EC)
into total dissolved solids (TDS).
Water type*
Ionic composition
HCO3-Ca
HCO3-Ca
HCO3-SO4-Ca
50% < HCO3-Ca < 70%
HCO3-SO4-Cl-Ca-Na
HCO3
SO4-HCO3-Cl-Ca-Na
Ca
SO4-Ca
SO4
Cl-HCO3-Na-Ca
Cl-Na
SO4-Ca (hyper SO4)
SO4
70%
SO4
Cl; Ca
50%
50%
70%; Ca
60%
50%
80%; Ca
70%
EC range
(dS/m)
FC
(mg/L)/(dS/m)
0.10 – 0.45
815
0.45 – 0.70
788
0.70 – 1.05
732
1.00 – 1.40
767
1.30 – 1.90
705
1.00 – 1.40
819
> 2.00
893
* Waters are named after Schoukarev’s criterion: the anions and cations that represent more
than 25% of the respective sums (in meq/L) are listed from higher to lower percentage.
Alternatively, the following two regression equations may be used to calculate
concentration in mg/L
TDS mg / L
749.0 66.3 EC 3.9 rCl 1.3 rHCO3 ; R 2
0.91; s 15.24 ,
where EC is the electrical conductivity in dS/m, and r is % of Cl or HCO3 over the sum of anions
(Cl+SO4+HCO3) in meq/L.
TDS mg / L
698.2 116.0 EC 55.6 SIgyp 99.9 (Cl/HCO3 ) 1.3 rNaK ; R 2
0.92; s 14.78
where EC is the electrical conductivity in dS/m, SIgyp is the saturation index for gypsum
[calculated with a geochemical model such as WATEQ4F], Cl/HCO3 is the ratio of Cl to HCO3 in
meq/L, and rNaK is the percentage (%) of sodium plus potassium over the sum of cations
(Na+K+Ca+Mg) in meq/L.
6. DESCRIPTION OF OUTPUTS
Figure 5 shows the “Output Data for the Area” menu and Table 11 summarizes the
names and acronyms of the main outputs. The outputs for CIRFLE are organized in three
columns: (1) water volume (Q) (given in the units specified in the “New Project” menu), (2) salt
concentration (C) (given in mg/L), and (3) salt mass (M) (given in Mg (mega-grams), Mg = t =
1000 kg). Some Parameters characterizing irrigation performance and salt outputs are given in
the small window to the right of the “Output Data for the Area” menu.
6.1. Water and salt fluxes
Each row of the “Output Data for the Area” menu gives the volume of water, the salt
concentration and the mass of salts associated to each model component. When the flow
components are in the vapor phase (Qevdiw, Qet and Qevp), C = M = 0 and the two right columns
are empty. Similarly, the salt fluxes originating from dissolution or resulting in precipitation (salt
pick-up – salt deposition and gypsum dissolution) are only shown as concentrations and
masses and the first column corresponding to the volume of water is empty.
It should be noted that Cisw is the salt concentration of the initial stored soil water once
the effect of gypsum, when present, is removed. Thus, if Gypsum = Yes, then Cisw = ECe (in
TDS units) - 2630 mg/L (the concentration of a gypsum-saturated solution).
The last row of the “Output Data for the Area” menu gives the three most important
outputs of CIRFLE: volume (Qsirf), salt concentration (Csirf) and mass of salts (Msirf) in surface
irrigation return flows. Qsirf results from the mixing of drainage waters (Qsdw), rim inflows (Qrim)
and surface runoff waters (Qiwro and Qpro) (Figure 1); Msirf is the sum of the mass of salts in
these flows (Figure 2), and Csirf is calculated as Msirf/Qsirf. As all the subcomponents of Qsirf and
Msirf are given in the screen, simple operations allow to obtain, for instance, the salt exports from
the irrigated soils only (Qrim excluded) or other particular results of interest.
All the intermediate Q, C and M components are also shown in this screen, allowing to
quantify for the most important sources of salts and its impact on drainage waters, irrigation
return flows, etc., for a particular simulation.
When clicking in the “Graphic” buttom of the “Output Data for the Area” menu, a
summary figure with the most important CIRFLE inputs and outputs for Q, C and M is shown
(Figure 6).
Table 11. Definition of intermediate and final variables in the Output data for the Area
menu (Q = volume of water; C = salt concentration; M = mass of salts)
Symbols .
Definition
Qdiw
Qeaiw
Cdiw
Ceaiw
Mdiw
Meaiw
Diverted irrigation water
Effective applied irrigation water
Qiwro
Ciwro
Miwro
Irrigation water runoff
Qevdiw
Evaporation of diverted irrigation water
Qp
Cp
Mp
Precipitation
Qep
Cep
Mep
Effective precipitation
Qpro
Cpro
Mpro
Precipitation runoff
Qevp
Evaporation of precipitation
Cse
Qisw
TDS of soil saturation extract corrected by gypsum solubility
Cisw
Misw
Qsw
Csw
Msw
Soil water after irrigation and precipitation/ before ET
Qpsw
Cpsw
Mpsw
Soil water after ET
Qet
Q’psw
Initial stored soil water
Actual Evapotranspiration
TDSgyp
Concentration derived from soil gypsum dissolution
Csp-Csd
Concentration derived from calcite pick-up/dissolution
C’psw
M’psw
C’fsw
Soil water after ET, salt pick-up/dissolution and gypsum dissolution
Final stored soil water corrected by gypsum
Qfsw
Cfsw
Mfsw
Final stored soil water
Qppsw
Cppsw
Mppsw
Water available for subsurface drainage and deep percolation
Qdp
Cdp
Mdp
Deep percolation
Qsdw
Csdw
Msdw
Subsurface drainage water
Qrim
Crim
Mrim
Rim inflow/outflow
Qsirf
Csirf
Msirf
Surface irrigation return flow
Figure 6. CIRFLE figure summarizing the most important inputs and outputs. For each of
them, the first line is volume of water (Q), the second line is salt concentration (C) and
the third line is mass of salts (M)
6.2. Irrigation performance and salt loading parameters
Some key parameters characterizing the quality of irrigation management and the
relative magnitude of salt fluxes are presented in the parameter output box to the right of the
output menu. When clicking on each of the parameters, a brief description is shown below the
box. The list of parameters is not exhaustive and the user may calculate other parameters of
interest (for his particular situation) from the output values in the main water and salt fluxes
window in the left.
Three parameters characterize the movement of water through the soil: the water use
efficiency (WUE), the ET concentration factor (ETCF) and the leaching fraction (LF).
The WUE is defined as the portion of the infiltrated irrigation and precipitation that
undergoes evapotranspiration
WUE
Q et
Q eaiw
Q ep
The ETCF is the ratio between the water in the soil “before” ET (Qsw) and “after” ET
(Qpsw) and gives an indication of how much the water in the soil gets concentrated due to the
evapotranspiration process (assumed salt free) during the simulated period
Q sw
Q psw
ETCF
Te leaching fraction (LF) is the ratio of the water draining below the root zone (Qppsw) to
the water flowing into the soil from irrigation and precipitation (Qeaiw + Qep).
LF
Q ppsw
Q eaiw
Q ep
Two parameters are calculated for the salt load of irrigation return flows: Mua and Muv.
The parameter Mua is the mass of salts per unit irrigated area (Mg TDS / irrigated hectare) and
is calculated as:
Mua (Mg / ha)
Msirf Mg
Area(ha)
The Mua is the mass of salts contributed by unit irrigated area (ha), possibly the most
important result form the model application. Strictly as defined, this parameter includes the
mass of salts carried by the rim inflows. When Qrim is important (or when the salt contribution of
the irrigated area alone is to be determined), the Mrim can be subtracted from the numerator.
Other modifications may be implemented likewise by the user from the main output data as
required.
The other parameter is the mass of salts per unit applied water volume (rainfall and
irrigation) calculated as:
Muv (Mg / hm 3
mg / L )
Msirf Mg
(Q diw
Q p )(hm 3 )
Again, if Mrim is not negligible, the contribution of the irrigated area to the salt load of the
irrigated area could be better assessed subtracting Mrim in the numerator and Qrim in the
denominator. This parameter has the units of concentration. It gives the concentration that result
from dissolving all the salts in the Area outflows (Msirf) in the volume of input water (Qp + Qdiw).
The comparison of this theoretical concentration with the input concentration (Cdiw or an
averaged mean of Cdiw and Cp) gives a hint of the salt removal from the irrigated soils, much in
the same way as the salt balance index (Kaddah and Rhoades, 1976). Neglecting minor terms,
if there were no gypsum or calcite dissolution in the soil (no loading) all the increase in salt
concentration in the drainage flow would result from evapoconcentration, and the ratio Muv/Cdiw
should be close to the ETCF. The difference between these terms is due first to the removal of
salts/gypsum from the soil and to the additional loading in surface flows and rim inflows.
This document was created with Win2PDF available at http://www.win2pdf.com.
The unregistered version of Win2PDF is for evaluation or non-commercial use only.
This page will not be added after purchasing Win2PDF.